[tahoe-dev] Benchmarks - 1.7.1 vs. 1.8.0c3

Kyle Markley kyle at arbyte.us
Tue Sep 7 05:08:56 UTC 2010


Zooko,
 
> Oh say it ain't so! Upgrading to 1.8.0c3 from 1.7.1 will make your
> large file downloads go about half as fast as they used to?

Don't despair -- my result might be true only for low-latency grids!  We
already have lots of data showing that 1.8.0c3 is an improvement for large
files on a real-world grid.

I think it would be useful to try an experiment like mine on the pubgrid. 
But I'm nervous about the storage consumed there.  We can't easily delete
that storage, so I'd have to always use the same convergence file.  But is
there any behavioral impact when uploading a file that already exists on
the grid?  (It wouldn't be in the backupdb because I'd always use a fresh
client, but the storage nodes would already have it.)

Or, with a few volunteers not behind NAT, we could move my testing into a
high-latency environment.  (Alas I've noticed the way I stop nodes isn't
totally reliable, so this might not be set-and-forget.)


> I really hope that this was some sort of glitch in your test rig.
> Could you please re-run this experiment a dozen times or so and post
> the results?

I'll run it continuously overnight and post the results.  But I'll scale
down from 40 storage nodes to 12; I think 10 nodes per machine was
excessive.

-- 
Kyle Markley


More information about the tahoe-dev mailing list